Many cytoplasmically-synthesized proteins must be correctly targeted to noncytoplasmic locations. The unifying feature between exported proteins in all systems is the requirement for a signal peptide. The principal objective of this work is to determine the structural requirements of signal peptides that are necessary for protein secretion and final localization. The overall aim is then to determine why these elements are important. For this purpose, a systematic series of mutants of the E. coli alkaline phosphatase gene will be produced. Each of these will be identical except for the amino acid composition of a subsegment of the amino-terminal signal peptide. In each mutant, this region will be designed to test the role of conformation, length, hydrophobicity, and overall topology. For example, we now have a series of mutants in the signal peptide hydrophobic core region which contain homopolymers and vary systematically in the hydrophobicity of the core region contituent residue and the length of this segment. This approach will be developed to include a series of mutants in the cleavage region designed to explore the structural features required for delivery to and recognition by the signal peptidase. A separate series of more global models will be constructed to explore the extent to which subsegments function as a unit and to determine the constraints for topological alignment and orientation. The rapid development of mutants, each containing multiple residue substitutions will be accomplished by development of the appropriate DNA restriction sites and cassette mutagenesis. This approach facilitates the comparison of entire structural units in well-defined parallel constructions. The ability of these mutants to support the delivery of alkaline phosphatase to the E. coli periplasm will be evaluated through in vivo studies. The relative dependence on an electrochemical potential, SecA and PrlA/SecY will be established. Transport defective mutants will be further evaluated for competence in membrane insertion, translocation, precursor processing and fidelity of the cleavage site. These studies are designed to reveal the specific signal sequence features critical at several different steps during the transport process. Isolated signal peptides corresponding to the wild type and mutant sequences will be chemically synthesized and purified to substantiate the physical character of each model. These will also be used for binding studies with components, such as SecA, likely to be involved in a given step during transport, with the aim of establishing the same hierarchy for binding in vitro as we observe for function in vivo. The structural features of signal peptides which enhance correct compartmentalization in E. coli will be useful for probing their eukaryotic counterparts. These principles can be applied to the tissue-specific targeting of therapeutic agents and the design of vehicles to transport other proteins, including eukaryotic proteins, into the E. coli periplasm for subsequent isolation.